The aim of this competitive renewal is to test an expanded causal model of patient and administrative outcomes, building upon the findings of the original NINR-funded grant - A Model of Patient and Nursing Administrative Outcomes. The expansion encompasses three domains. First, the theoretical model will be expanded by incorporating additional relevant contextual, structural and outcome variables that have been suggested by findings of recent research and the original project. Second, several key variables (professional practice, staffing adequacy, and patient outcomes) will be reconceptualized and their measurement refined. Third, generalizability of the model will be improved by moving from a regional to a national random sample. The research questions to be addressed are: 1. To what extent to contextual characteristics (external environment, organizational environment, internal environment) affect staffing adequacy and professional practice? 2. To what extent does staffing adequacy affect professional practice? 3. What are the effects of professional practice on administrative outcomes (nurses' job satisfaction, nursing turnover, cost, and average length of stay) and patient outcomes (patient satisfaction, management of symptom distress, adverse events)? 4. What are the effects of staffing adequacy on administrative and patient outcomes? 5. How does the hospital level structural equation model differ from the nursing unit structural equation model? The design is a non-experimental causal modeling study in which the nursing unit is the unit of analysis; data will be collected over a six month period from hospital study coordinators, staff nurses, and patients. The sample is a national random sample of 160 general acute care hospitals with more than 150 beds; two general medical-surgical units in each hospital will participate. The analysis will use multilevel structural equation modeling, which incorporates latent variables and provides: correct standard errors for the cluster sample; simultaneous tests of predictors of dependent variables at different hierarchical levels; and goodness of fit statistics. The significance of the project lies in its ability to refine knowledge about the relationships of professional nursing practice, staffing adequacy and outcomes, which will contribute to development system-level administrative interventions to improve care.